11 August 2015

Acknowledgements

  • Tom Whitham (Northern Arizona University)
  • Stuart Borrett (UNC Wilmington)
  • Stephen Shuster (Northern Arizona University)
  • Jamie Lamit (Michigan Technical University)
  • Todd Wojtowicz (Northern Arizona University)
  • Art Keith (Northern Arizona University)
  • David Solance Smith (Denison College)
  • Matt Zinkgraf (UC Davis)
  • Luke Evans (West Virginia University)
  • Rikke Naesborg (UC Berkeley)
  • bipartite (Dormann et al. 2008)

Take Home Slide

  • Ecological and evolutionary processes are linked.
  • Eco-evo investigations have focused on species pairs, while network ecology has little genetics.
  • We investigated how genetically based variation in a foundation species impacts the structure of ecological networks.
  • Take-home: genetic variation in a foundation species can impact ecological network structure, which could influence the dynamics and evolution of complex communities.

The interface of ecology and evolution

Community genetics is important

Bailey et al. 2012 New Phytologist

Why study networks?

Why study networks?

Alligator

Why study networks?

Alligator

Bondavalli and Ulanowicz 1999 Ecosystems

Why study networks?

Alligator

Bondavalli and Ulanowicz 1999 Ecosystems

Why study networks?

Alligator

Bondavalli and Ulanowicz 1999 Ecosystems

Why study networks?

Alligator

Bondavalli and Ulanowicz 1999 Ecosystems

Why study evolution of networks?

Trophic cascades across ecosystems

Knight et al. 2005 Nature

Modularity

utah from outerspace

Fortuna et al. 2009

Utah from outer-space

utah from outerspace

Utah at ground-level

common garden

cottonwood tree

cottonwood tree

insects

Keith et al. 2010 Ecology

cottonwood tree

insects

Genotype-Species Network

Figure 1. Cottonwood genotype and arthropod species network

Lau et al. (submitted Ecology)

Genotype-Species Network

Figure 1. Cottonwood genotype and arthropod species network

  • Modular network structure (z =3.82, P = 0.038)

Lau et al. (submitted Ecology)

Simulated communities

Simulated communities

Warning - Math!

Simulated communities

Shuster et al. 2006

\[ n^{*}_{ij} = K [1 - \frac{\gamma}{2} \sigma^2_{z_{ij}} - \frac{\gamma}{2} (\theta_i - \bar{z}^{*}_{ij})] + E_{n_{ij}} \]

Simulated communities

Shuster et al. 2006

\[ n^{*}_{ij} = K [1 - \frac{\gamma}{2} \sigma^2_{z_{ij}} - \frac{\gamma}{2} (\theta_i - \bar{z}^{*}_{ij})] + E_{n_{ij}} \]

leaf

Simulated communities

Shuster et al. 2006

\[ n^{*}_{ij} = K [1 - \frac{\gamma}{2} \sigma^2_{z_{ij}} - \frac{\gamma}{2} (\theta_i - \bar{z}^{*}_{ij})] + E_{n_{ij}} \]

leaf insect proboscis

Simulated communities

Shuster et al. 2006

\[ n^{*}_{ij} = K [1 - \frac{\gamma}{2} \sigma^2_{z_{ij}} - \frac{\gamma}{2} (\theta_i - \bar{z}^{*}_{ij})] + E_{n_{ij}} \]

insects

Simulated communities

Figure 2. simulations

Lau et al. (submitted Ecology)

Genotypes -> network structure

Figure 3. simulations zoomed

Lau et al. (submitted Ecology)

Lichen Interaction Networks

cottonwood tree

Lichen Interaction Networks

cottonwood tree

cottonwood tree

Lichen - Garden and Wild

Figure 2. Genotypic variation generates interaction network structure

Lau et al. (In Prep)

Lichen - Garden and Wild

Figure 3. Genotypic variation generates interaction network structure

Lau et al. (In Prep)

Conclusions

All models are wrong: some are useful…

darwin

All models are wrong: some are useful, some are not.

darwin

Selection: Impacts on Networks

modes of selection

Connor & Hartl 2004 A Primer of Ecological Genetics

(Human)-Assisted Evolution

assisted evolution

  • Van Oppen et al. 2015 PNAS

Thanks!

Acknowledgements

  • Tom Whitham (Northern Arizona University)
  • Stuart Borrett (UNC Wilmington)
  • Stephen Shuster (Northern Arizona University)
  • Jamie Lamit (Michigan Technical University)
  • Todd Wojtowicz (Northern Arizona University)
  • Art Keith (Northern Arizona University)
  • David Solance Smith (Denison College)
  • Matt Zinkgraf (UC Davis)
  • Luke Evans (West Virginia University)
  • Rikke Naesborg (UC Berkeley)
  • bipartite (Dormann et al. 2008)